<feed xmlns='http://www.w3.org/2005/Atom'>
<title>glusterfs.git/rpc/rpc-transport/rdma/src/rdma.h, branch v3.1.0qa34</title>
<subtitle></subtitle>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/'/>
<entry>
<title>rpc/rpc-transport: fix frame-loss during rdma-reads.</title>
<updated>2010-09-24T14:10:27+00:00</updated>
<author>
<name>Raghavendra G</name>
<email>raghavendra@gluster.com</email>
</author>
<published>2010-09-24T07:51:53+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=6e952607f17dd5ea6e3748f2bbc26234f4b1e553'/>
<id>6e952607f17dd5ea6e3748f2bbc26234f4b1e553</id>
<content type='text'>
- total no of rdma reads in progress was tracked using refcount of post
    (refcount = total rdma reads + 1). However doing rdma_post_unref _after_
    rdma_do_reads, can result in race condition between execution of
    rdma_post_unref and procedure handling rdma read completion. This makes
    it impossible to find out whether the current rdma read was the last
    rdma-read being done in procedure handling rdma read-completion.
    The fix to this should either do rdma_post_unref before posting an rdma
    read or use another variable to track the number of rdma-reads.
    This patch implements second method.

Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 1700 (write fops are bailing out on a distribute setup)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1700
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
- total no of rdma reads in progress was tracked using refcount of post
    (refcount = total rdma reads + 1). However doing rdma_post_unref _after_
    rdma_do_reads, can result in race condition between execution of
    rdma_post_unref and procedure handling rdma read completion. This makes
    it impossible to find out whether the current rdma read was the last
    rdma-read being done in procedure handling rdma read-completion.
    The fix to this should either do rdma_post_unref before posting an rdma
    read or use another variable to track the number of rdma-reads.
    This patch implements second method.

Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 1700 (write fops are bailing out on a distribute setup)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=1700
</pre>
</div>
</content>
</entry>
<entry>
<title>rpc-transport/rdma: change the default listening port to 6968.</title>
<updated>2010-09-08T09:36:17+00:00</updated>
<author>
<name>Raghavendra G</name>
<email>raghavendra@gluster.com</email>
</author>
<published>2010-09-08T01:20:43+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=d2e2e69bb4a212df08c3c4469cb91a88d244ed9f'/>
<id>d2e2e69bb4a212df08c3c4469cb91a88d244ed9f</id>
<content type='text'>
Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 513 (Introduce 0 copy rdma)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=513
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 513 (Introduce 0 copy rdma)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=513
</pre>
</div>
</content>
</entry>
<entry>
<title>Bring in new transport rdma.</title>
<updated>2010-08-30T13:54:12+00:00</updated>
<author>
<name>Raghavendra G</name>
<email>raghavendra@gluster.com</email>
</author>
<published>2010-08-30T08:03:52+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=300b4fefcbb5a5ced4f0554e109679e44cdf44ea'/>
<id>300b4fefcbb5a5ced4f0554e109679e44cdf44ea</id>
<content type='text'>
- rdma is new transport and improvement over current ib-verbs. It uses
   rdma-read and rdma-write for efficiently transferring large buffers.
   For more details please refer to rfc-5666 and rfc-5667.

Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 513 (Introduce 0 copy rdma)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=513
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
- rdma is new transport and improvement over current ib-verbs. It uses
   rdma-read and rdma-write for efficiently transferring large buffers.
   For more details please refer to rfc-5666 and rfc-5667.

Signed-off-by: Raghavendra G &lt;raghavendra@gluster.com&gt;
Signed-off-by: Vijay Bellur &lt;vijay@dev.gluster.com&gt;

BUG: 513 (Introduce 0 copy rdma)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=513
</pre>
</div>
</content>
</entry>
</feed>
